Chapter 2

A Free-Air Gravity Anomaly Map of Long Island and Block Island Sounds

Peter Dehlinger1, Lanbo Liu2 and Ralph S. Lewis3
Table of Contents
Marine Gravity Data Acquisition Procedures
Data Reduction Procedures
Error Analysis
Gravity Anomaly Discussions
Figure Captions
Table Captions
Appendix I: Some background information on the first author
Digital Data and Metadata
Back to Table of Contents


The late Dr. Peter Dehlinger, a Professor in the Department of Geology and Geophysics of the University of Connecticut (Appendix I), conducted marine gravity surveys in the Long Island and Block Island Sounds in early 1970s. Two free-air gravity anomaly maps: one for the Long Island Sound and the other for the Block Island Sound have appeared only as separate figures (in a textbook) at a very small scale. These data are valuable in many ways. The most important role of these data is that they provide a geological framework for further investigations of the marine environment. Based on Dr. Dehlinger's original files and written communications, as well as updated global gravity knowledge, a new free-air gravity anomaly map is provided here as a single map containing both the Long Island Sound and Block Island Sound. This paper provides documentation for this map, which was digitized from original blue prints. The contour interval of the map is 2 mgal since the estimated error is also at the 2-mgal level. Data reduction procedures and accuracy are discussed. The geological implications of interesting anomalies are also briefly discussed.


The Long Island Sound (LIS) is an estuary located between Long Island, in the state of New York, and the state of Connecticut. It opens to Block Island Sound (BIS) to the east, which, in turn, opens to the Atlantic Ocean. The two Sounds are interconnected and are underlain by continental crust. The region has been tectonically inactive since the Triassic Basin of Connecticut's Central Valley was formed along a zone of local weakness in the crust, between two semi-parallel chains of older gneiss domes (Bell, 1985). In recent years new geological and geophysical surveys have been conducted in LIS (Poppe and Polloni, 1998). Nevertheless, no new gravity surveys have been made since the 1975 surveys (Dehlinger, unpublished data available at Department of Geology and Geophysics, University of Connecticut). The gravity data are unique and can be used to complement other subsurface investigations. The present paper is a summary of information based on old manuscripts, correspondences among authors and reviewers, as well as communication with professionals who had been involved in the marine gravity survey in the Long Island and Block Island Sounds. The purpose is to disseminate existing gravity information to a broader research community, and to facilitate modern geological and environmental studies of the Sounds.


The University of Connecticut made marine gravity measurements in LIS and BIS in 1975. Free-air gravity anomalies were initially calculated relative to the 1930 International Gravity Formula and the original Potsdam value of absolute gravity. In a later study, the free-air anomalies were recalculated by Dr. Dehlinger relative to the 1967 International Gravity Formula (Moritz, 1968) and the revised Potsdam base value, resulting in revised free-air anomaly maps (1:80,000 scale) for both LIS and BIS. The resultant free-air anomaly maps were included in his textbook 'Marine Gravity', but only on paper and at a very small scale (Dehlinger, 1978). Although these maps were not published at a usable scale during Dr. Dehlinger's life, it was his wish that these maps be published to reach a broader research community.

The gravity surveys were conducted aboard the University of Connecticut 65-foot T-Boat (since named R/V UCONN). Gravity was continuously measured by a platform-mounted LaCoste & Romberg gravimeter (No. S-71). The sensing element on this gravimeter is dampened and constrained to the vertical direction making it Figure 1 - Tracklinespossible to achieve high sensitivities without concern for possible motion related to sea state. Gravity measurements were obtained at 1.5- to 2.5-km intervals (approximately 15-minute time intervals) along the constant-speed ship-track lines shown in Figure 1. Measurements were made in calm seas only, where ship accelerations did not significantly influence attempts to measure gravity accelerations.

Ship positions were obtained by visual sight and radar fixes to nearby shore features that were clearly identifiable on navigation charts of the Sounds (LORAN C had not been installed at the time of the surveys). Fixes were obtained as frequently as they could be established, usually at distances of l to 15 km along track lines. Final track lines shown on the map (Fig. 1) are smoothed lines through the plotted fixes, and actual ship positions were obtained from interpolations along these lines. The resultant ship positions and corresponding gravity measurements produced mean gravity uncertainties of ± 2.1 mgal at track line intersections. These uncertainties indicate a mean error in gravity measurement of ± 1.5 mgal. Nevertheless, based on the error analysis we conducted (see Error Analysis section) the realistic estimate of the uncertainty should be around 2 mgal.


A gravity base station was established at the ship's homeport, the Marine Laboratory of the University of Connecticut at Noank, Connecticut (Table 1). The value of gravity at this station is 980273.9 mgal, relative to the revised Potsdam base value of 981260 mgal. The Noank base station value was determined by a geodetic land gravity meter relative to known gravity at the following four base stations in Connecticut. (1) the Hartford-Springfield Bradley Airport Station; (2) the Tweed-New Haven Airport Station; (3) the Yale University Station; and (4) the U. S. Submarine Base Station. Gravity Base Stations for locations are given in Table 1.

Eotvos corrections (Dehlinger 1978, Equation 6.43) were calculated for each gravity measurement using the formula

    C = 7.5 vE cos ø (mgal)
in which vE is the easterly component of ship speed, in knots, and ø the geographic latitude. The final adjusted track lines provided data used to obtain ship speed and latitude.

The final contours were merged with contours of land data in Connecticut and Rhode Island (Bothner et al, 1978; Urban et al, 1972; Hildreth, 1979), and with anomalies on Long Island.


One major source of error in the free-air gravity anomaly contour map comes from the positioning error of the survey vessel. The navigation is poorly constrained (only using radar and visual fixes). During the surveys made in LIS, usually four visual or radar sights along prominent coastal features were obtained for one navigation fix while the ship was underway. At least in LIS, this means the ship was near the middle of the intersecting lines of sight. In BIS, the fixes were constrained at an even lower level compared to that in LIS. The navigation fixes were generally within a quarter of a mile of the adjusted positions. However, since the gravity gradients for most of  LIS are small, with a few exceptional areas, it is reasonably easy to draw contour lines at the 2-mgal contour interval.

The original 1:80,000-scale maps depict one-mgal contour intervals in the context of the argument that ± 1.5 mgal mean accuracy had been achieved. After careful examination of the error propagation path, we suggest that a level of 2-mgal accuracy might be a more realistic estimation. The mean line-crossing error is about 2.1 mgal. Thus, using the 2-mgal interval instead of the original 1-mgal interval is probably more appropriate. Though the digital data are not accessible, digitizing the gravity anomaly contours from the original 1:80,000-scale blue prints proved feasible and this process was used to produce the map presented here.


Figure 2 - Gravity anomiliesThe regional free-air gravity map of LIS and BIS is presented in Figure 2.  To open a georeferenced display of this theme in ESRI's ArcView program make sure the application is loaded on your computer.  Users should go to the lisound directory located on the top level of this CD-ROM and double click on the lisound.apr project file.  The individual ArcView shapefiles may also be opened directly with any Arc application (e.g. ArcInfo, ArcExplorer) and can also be found on the data page.  Further detailed information can be found on the ArcView Project File page.

The general trend of the free-air anomaly map shows positive values in the western end of the LIS, with a background value about +20 mgal (Fig. 2). The anomaly gradually decreases to negative values toward the east, with a background value of -10 mgal in the BIS. This is the anomaly variation that would be expected across a continental margin, where changes in crustal thickness and density occur approaching the oceanic crusts (Dehlinger and Jones, 1965). In the western part of LIS, a prominent feature on the anomaly maps is the N-S trending gravity high extending from Greenwich, CT to Lloyd Neck on Long Island, NY. This is the Appalachian Gravity High that extends continuously across New Jersey, Long Island, and western Connecticut (Woollard, 1943; Longwell, 1943).  The maximum free-air value of this high is +45 mgal in LIS (long. 73d 24' W). The gradients on each side of the high are -2.5 mgal/km for about 10 km on the west side, and -1 mgal/km for about 35 km on the east side. The gravity field in the eastern half of the LIS map consists of long wavelength, low amplitude anomalies, superimposed on a regional westerly gradient. The typical values range from about -10 mgal on the east side (longitude 72d 20' W) to about +10 mgal on the west side (longitude 72d 55' W).

Anomalies in BIS have generally shorter wavelengths than those in eastern LIS. In the northern half of BIS, the gravity field has values near -10 to -15 mgal; southward the gravity field increases at about 2 mgal/km, to values of about +5 mgal in the southern part of the Sound. Several NW-SE trending anomalies are superimposed on the regional BIS gravity field.

The anomaly maps provide a reasonable estimate of the free-air field in the water- covered area between land-based Bouguer anomaly maps across New England and Long Island (Bothner and others, 1978; Urban and others, 1972; Hildreth, 1979).


In the data acquisition phase, Dr. E. F. Chiburis and several of the graduate students in geophysics at the University of Connecticut assisted with the data reduction and established the Noank gravity base station. Captain John Blume obtained the navigation fixes aboard ship. Karen Kastening drafted the original anomaly maps. We are in debt to Dr. Thomas Hildenbrand of USGS for his careful reviews and constructive comments to the early drafts of this paper. Mary DiGiacomo-Cohen digitized the map. Digital copies of the data and maps presented here are available at University of Connecticut.


Bell, M., 1985, The face of Connecticut: people, geology, and the land: Connecticut Geological and Natural History Survey, 196 p.

Bothner, W. A., Bromery, R. W., Davis, M., and Ahmad, F., 1978, Preliminary complete Bouguer gravity anomaly map of Connecticut: U. S. Geological Survey Open-File Report 78-816, scale 1:250,000.

Dehlinger, P., and Jones, B. R., 1965, Free-air gravity anomaly map of the Gulf of Mexico and its tectonic implications: Geophysics, v. 30, no. 1, p. 102-110.

Dehlinger, P., 1978, Marine gravity: New York, Amsterdam, Elsevier, 322 p.

Dobrin, M.B., 1960, Introduction to Geophysical Prospecting: New York, McGraw-Hill Book Company, 446 p.

Fowler, C.M.R., 1990, The Solid Earth: New York, Cambridge University Press, 472 p.

Hildreth, C. T. (Compiler), 1979, Bouguer gravity map of northeastern United States and southeastern Canada, offshore and onshore: New York State Museum Map and Chart Series no. 32, scale 1:1,000,000.

Longwell, C. R., 1943, Geologic interpretation of gravity anomalies in the southern New England-Hudson Valley region: Geological Society of America Bulletin, v. 54, p. 555-590.

Moritz, H., 1968, The geodetic reference system 1967: Allgemeine Vermessungs Nachrichten, v. 75, p. 2-7.

Poppe, L. J., and Polloni, C., 1998, Long Island Sound environmental studies: US Geological Survey Open-File Report 98-502.

Press, F. and Siever, R., 1974, Earth: New York, W.H. Freeman and Company, 656 p.

Urban, T. C., Bromery, R. W., Revetta, F. A., and Diment, W. H., 1972, Simple Bouguer gravity anomaly map of southeastern New York and contiguous states: New York Museum and Science Service, Geological Survey Map and Chart Series no. 17C, scale 1:250,000.

Woollard, G. P., 1943, Geologic correlation of areal gravitational and magnetic studies of New Jersey and vicinity: Geological Society of America Bulletin, v. 54, p. 791-818.

Woollard, G. P., and Rose, J. C., 1963, International Gravity Measurements: Society of Exploration Geophysicists, Tulsa, Oklahoma, 518 p.


Figure 1. The ship track lines for gravity surveys in Long Island and Block Island Sounds. The positions were determined by a combination of visual sight and radar fixes to nearby shore features.

Figure 2.  Small-scale map of the free-air gravity anomalies in the Long Island and Block Island sounds. The original maps for the Long Island Sound and the Block Island sound were separate at the N-S edge through Groton, Connecticut.


Table 1.  List of gravity base stations in located Connecticut and used during this study.  Gravity values are relative to the revised Potsdam base value of 981260 mgal.

1 deceased, Department of Geology and Geophysics, University of Connecticut, Storrs, CT 06269
2 Department of Geology and Geophysics, University of Connecticut, Storrs, CT 06269
3 Geological and Natural History Survey Connecticut Department of Environmental Protection, Hartford, CT 06106 [an error occurred while processing this directive]